Method and system to predict one or more trajectories of a vehicle based on context surrounding the vehicle

Abstract

A surrounding environment of an autonomous vehicle is perceived to identify one or more vehicles nearby. For each of the identified vehicles, based on a current location of the identified vehicle, vehicle-independent information is obtained to determine context surrounding the identified vehicle, where the vehicle-independent information includes vehicle surrounding information that defines physical constraints imposed on the identified vehicle. For each of the identified vehicles, one or more trajectories for the identified vehicle are predicted based at least in part on the vehicle-independent information associated with the identified vehicle. The autonomous vehicle is controlled based on the one or more predicted trajectories of the one or more identified vehicles.

What is claimed is: 
     
       1. A computer-implemented method for operating an autonomous vehicle, the method comprising:
 perceiving a surrounding environment of the autonomous vehicle to identify one or more vehicles nearby; 
 for each of the identified vehicles,
 based on a current location of the identified vehicle, obtaining vehicle-independent information, including physical constraints imposed on the identified vehicle, 
 predicting a first set of one or more trajectories for the identified vehicle by applying a first machine learning model to a first set of features extracted from the vehicle-independent information associated with the identified vehicle, 
 predicting a second set of one or more trajectories by applying a second machine learning model to a second set of features extracted from sensor data captured by one or more sensors of the autonomous vehicle perceiving behaviors of the identified vehicle, and 
 determining a final set of predicted trajectories based on the first set of trajectories and the second set of trajectories; and 
 
 controlling the autonomous vehicle based on one or more final predicted trajectories of the one or more identified vehicles. 
 
     
     
       2. The method of  claim 1 , further comprising
 for each of the identified vehicles,
 determining the current location of the identified vehicle based at least in part on the sensor data. 
 
 
     
     
       3. The method of  claim 1 , wherein the vehicle-independent information further includes a time when the vehicle is identified, driving conditions, points of interest (POI) and event information, and traffic information that further impose the physical constraints on the vehicle. 
     
     
       4. The method of  claim 3 , wherein the POI and event information includes information representing a destination heading for the identified vehicle. 
     
     
       5. The method of  claim 1 , wherein the first machine learning model is configured to output the first set of one or more trajectories based on the vehicle-independent information. 
     
     
       6. The method of  claim 1 , wherein determining a final set of predicted trajectories comprises:
 merging the first set of predicted trajectories and the second set of predicted trajectories to obtain the final set of predicted trajectories. 
 
     
     
       7. The method of  claim 6 , wherein merging the first set of predicted trajectories and the second set of predicted trajectories to obtain the final set of predicted trajectories is performed using a Bayesian algorithm. 
     
     
       8. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for operating an autonomous vehicle, the operations comprising:
 perceiving a surrounding environment of the autonomous vehicle to identify one or more vehicles nearby; 
 for each of the identified vehicles,
 based on a current location of the identified vehicle, obtaining vehicle-independent information, including physical constraints imposed on the identified vehicle, 
 predicting a first set of one or more trajectories for the identified vehicle by applying a first machine learning model to a first set of features extracted from the vehicle-independent information associated with the identified vehicle, 
 predicting a second set of one or more trajectories by applying a second machine learning model to a second set of features extracted from sensor data captured by one or more sensors of the autonomous vehicle perceiving behaviors of the identified vehicle, and 
 determining a final set of predicted trajectories based on the first set of trajectories and the second set of trajectories; and 
 
 controlling the autonomous vehicle based on one or more final predicted trajectories of the one or more identified vehicles. 
 
     
     
       9. The machine-readable medium of  claim 8 , wherein the operations further comprise:
 for each of the identified vehicles,
 determining the current location of the identified vehicle based at least in part on the sensor data. 
 
 
     
     
       10. The machine-readable medium of  claim 8 , wherein the vehicle-independent information further includes a time when the vehicle is identified, driving conditions, points of interest (POI) and event information, and traffic information that further impose the physical constraints on the vehicle. 
     
     
       11. The machine-readable medium of  claim 10 , wherein the POI and event information includes information representing a destination heading for the identified vehicle. 
     
     
       12. The machine-readable medium of  claim 8 , wherein the first machine learning model is configured to output the first set of one or more trajectories based on the vehicle-independent information. 
     
     
       13. The machine-readable medium of  claim 8 , wherein determining a final set of predicted trajectories comprises:
 merging the first set of predicted trajectories and the second set of predicted trajectories to obtain the final set of predicted trajectories. 
 
     
     
       14. The machine-readable medium of  claim 13 , wherein merging the first set of predicted trajectories and the second set of predicted trajectories to obtain the final set of predicted trajectories is performed using a Bayesian algorithm. 
     
     
       15. A data processing system, comprising:
 a processor; and 
 a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for operating an autonomous vehicle, the operations including:
 perceiving a surrounding environment of the autonomous vehicle to identify one or more vehicles nearby; 
 for each of the identified vehicles,
 based on a current location of the identified vehicle, obtaining vehicle-independent information, including physical constraints imposed on the identified vehicle, 
 
 predicting a first of one or more trajectories for the identified vehicle by applying a first machine learning model to a first set of features extracted from the vehicle-independent information associated with the identified vehicle,
 predicting a second set of one or more trajectories by applying a second machine learning model to a second set of features extracted from sensor data captured by one or more sensors of the autonomous vehicle perceiving behaviors of the identified vehicle, and 
 determining a final set of predicted trajectories based on the first set of trajectories and the second set of trajectories; and 
 
 controlling the autonomous vehicle based on one or more final predicted trajectories of the one or more identified vehicles. 
 
 
     
     
       16. The system of  claim 15 , wherein the operations further include:
 for each of the identified vehicles,
 determining the current location of the identified vehicle based at least in part on the sensor data. 
 
 
     
     
       17. The system of  claim 15 , wherein the vehicle-independent information further includes a time when the vehicle is identified, driving conditions, points of interest (POI) and event information, and traffic information that further impose the physical constraints on the vehicle. 
     
     
       18. The system of  claim 17 , wherein the POI and event information includes information representing a destination heading for the identified vehicle. 
     
     
       19. The machine-readable medium of  claim 15 , wherein the first machine learning model is configured to output the first set of one or more trajectories based on the vehicle-independent information. 
     
     
       20. The system of  claim 15 , wherein determining a final set of predicted trajectories comprises:
 merging the first set of predicted trajectories and the second set of predicted trajectories to obtain a final set of predicted trajectories. 
 
     
     
       21. The system of  claim 20 , wherein merging the first set of predicted trajectories and the second set of predicted trajectories to obtain the final set of predicted trajectories is performed using a Bayesian algorithm.